These are the errors that I came across while I was trying to preprocess the datasets:
python preprocess.py -train_src data/train/article.txt -train_tgt data/train/summary.txt -valid_src data/train/article.txt -valid_tgt data/train/summary.txt -save_data data/train/t Building source vocabulary... Created dictionary of size 4227 (pruned from 4227) () Building target vocabulary... Created dictionary of size 675 (pruned from 675) () Preparing training ... Processing data/train/article.txt & data/train/summary.txt ... ... shuffling sentences Traceback (most recent call last): File "preprocess.py", line 237, in <module> main() File "preprocess.py", line 216, in main dicts['src'], dicts['tgt']) File "preprocess.py", line 184, in makeData perm = torch.randperm(len(src)) RuntimeError: must be strictly positive at /home/ubuntu/pytorch/torch/lib/TH/generic/THTensorMath.c:1984
Kindly, let me know the solution for this. Thank you in advanced.